# Time Imprint: Learning Time-Aware Representations in Multi-Modal Knowledge Graphs

arXiv: https://arxiv.org/abs/2607.09777 · HTML: https://mlsanity.com/paper/2607.09777/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 25 of its 44 references that are in the map.
- First posted on arXiv 8 Jul 2026.

## Builds on (most cited first, 25 of 25)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Embedding Entities and Relations for Learning and Inference in Knowledge Bases](https://mlsanity.com/paper/1412.6575/) (2014)
- [Complex Embeddings for Simple Link Prediction](https://mlsanity.com/paper/1606.06357/) (2016)
- [RotatE: Knowledge Graph Embedding by Relational Rotation in Complex Space](https://mlsanity.com/paper/1902.10197/) (2019)
- [BEiT v2: Masked Image Modeling with Vector-Quantized Visual Tokenizers](https://mlsanity.com/paper/2208.06366/) (2022)
- [Temporal Knowledge Graph Reasoning Based on Evolutional Representation Learning](https://mlsanity.com/paper/2104.10353/) (2021)
- [MMKG: Multi-Modal Knowledge Graphs](https://mlsanity.com/paper/1903.05485/) (2019)
- [Diachronic Embedding for Temporal Knowledge Graph Completion](https://mlsanity.com/paper/1907.03143/) (2019)
- [Learning from History: Modeling Temporal Knowledge Graphs with Sequential Copy-Generation Networks](https://mlsanity.com/paper/2012.08492/) (2020)
- [Multi-Modal Knowledge Graph Construction and Application: A Survey](https://mlsanity.com/paper/2202.05786/) (2022)
- [Image-embodied Knowledge Representation Learning](https://mlsanity.com/paper/1609.07028/) (2016)
- [Inductive Entity Representations from Text via Link Prediction](https://mlsanity.com/paper/2010.03496/) (2020)
- [Visual Pivoting for (Unsupervised) Entity Alignment](https://mlsanity.com/paper/2009.13603/) (2020)
- [A Survey on Temporal Knowledge Graph: Representation Learning and Applications](https://mlsanity.com/paper/2403.04782/) (2024)
- [IMF: Interactive Multimodal Fusion Model for Link Prediction](https://mlsanity.com/paper/2303.10816/) (2023)
- [NativE: Multi-modal Knowledge Graph Completion in the Wild](https://mlsanity.com/paper/2406.17605/) (2024)
- [Multi-modal Entity Alignment in Hyperbolic Space](https://mlsanity.com/paper/2106.03619/) (2021)
- [Knowledge Graph Completion with Pre-trained Multimodal Transformer and Twins Negative Sampling](https://mlsanity.com/paper/2209.07084/) (2022)
- [Tokenization, Fusion, and Augmentation: Towards Fine-grained Multi-modal Entity Representation](https://mlsanity.com/paper/2404.09468/) (2024)
- [Multiple Heads are Better than One: Mixture of Modality Knowledge Experts for Entity Representation Learning](https://mlsanity.com/paper/2405.16869/) (2024)
- [Noise-powered Multi-modal Knowledge Graph Representation Framework](https://mlsanity.com/paper/2403.06832/) (2024)
- [Progressively Modality Freezing for Multi-Modal Entity Alignment](https://mlsanity.com/paper/2407.16168/) (2024)
- [End-to-End Learning on Multimodal Knowledge Graphs](https://mlsanity.com/paper/2309.01169/) (2023)
- [Multi-modal Semantic Understanding with Contrastive Cross-modal Feature Alignment](https://mlsanity.com/paper/2403.06355/) (2024)
- [Transformer-Based Multimodal Knowledge Graph Completion with Link-Aware Contexts](https://mlsanity.com/paper/2501.15688/) (2025)

## Abstract (arXiv)

Multi-Modal Knowledge Graphs (MMKGs) enrich entities with multiple modalities such as text and images, yet entities with highly similar multi-modal features remain difficult to distinguish. Temporal information of an entity can serve as an additional modality to disambiguate such entities, but existing approaches rarely treat time as a separate modality alongside text and images due to two major challenges: (1) sparse temporal semantics, which hinder alignment with richer modalities, and (2) multiple timestamps, which introduce noise or reduce robustness in representation learning. To address these challenges, we propose Time Imprint, a framework that treats time as an entity-level modality and jointly aligns temporal, textual, and visual representations via a three-view contrastive objective. Additionally, to mitigate multi-timestamp ambiguity, Time Imprint studies a compact timestamp subset selection design space and aggregates the selected timestamps into a discriminative temporal embedding with attention pooling, balancing temporal specificity and robustness. Experiments on three MMKG benchmarks demonstrate that Time Imprint achieves state-of-the-art link prediction performance, improving Hits@1 by up to 6.07% overall and yielding up to 58% gains on the subset of the top-1% ambiguity samples. We further examine different fusion strategies and the sensitivity to timestamp availability and quality, clarifying when and why time-as-modality is most beneficial, while adding only modest training overhead. We release our code at https://anonymous.4open.science/r/Time-Imprint.